We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Large Language Models are prone to biased predictions and hallucinations, underlining the paramount importance of understanding their model-internal reasoning process. However, achieving faithful attributions for the entirety of a black-box transformer model and maintaining computational efficiency is an unsolved challenge. By extending the Layer-wise Relevance Propagation attribution method to handle attention layers, we address these challenges effectively. While partial solutions exist, our method is the first to faithfully and holistically attribute not only input but also latent representations of transformer models with the computational efficiency similar to a single backward pass. Through extensive evaluations against existing methods on LLaMa 2, Mixtral 8x7b, Flan-T5 and vision transformer architectures, we demonstrate that our proposed approach surpasses alternative methods in terms of faithfulness and enables the understanding of latent representations, opening up the door for concept-based explanations. We provide an LRP library at https://github.com/rachtibat/LRP-eXplains-Transformers.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2402.05602")
get_code_for_paper("2402.05602")
have("2402.05602")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2402.05602.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2402.05602)
Connect an agent — have() is free.